摘要目的:探讨胆囊癌根治术中淋巴结清扫数目对患者预后的影响。方法:回顾性分析2013年1月至2017年12月中国6家医疗中心收治的401例行意向性根治性手术的胆囊癌患者的临床病理学资料和随访资料。其中男性153例（38.2%），女性248例（61.8%），年龄（62.0±10.5）岁（范围：30~88岁）。53例（22.2%）患者术前伴有黄疸症状。所有患者均接受意向性根治性切除+区域淋巴结清扫，术后病理学检查结果证实为R0或R1切除。采用X-tile软件以不同淋巴结清扫数目作为截断值，分组进行统计检验，分析不同截断值与预后的关系，寻找最佳截断值。生存分析采用Kaplan-Meier法，单因素预后分析采用Log-rank检验，多因素预后分析采用Cox比例风险模型。结果:401例胆囊癌患者中，135例（33.6%）伴有淋巴结转移，其中N1期98例（24.4%），N2期37例（9.2%）。共清扫2 794枚淋巴结，每例患者淋巴结清扫数目[ M（ Q R）]为6（5）枚，阳性淋巴结数目为0（1）枚，阳性淋巴结比例为0（0.20）。X-tile软件分析结果显示，淋巴结清扫数目12枚和15枚为两个界值，据此将全部患者按淋巴结清扫数目分为1~11枚、12~15枚、≥16枚三组，3年累积生存率分别为45.2%、74.5%、12.0%，三组间差异有统计学意义（χ 2=10.94， P<0.01）。多因素分析结果显示，淋巴结清扫数目是胆囊癌患者的独立预后因素（ P<0.05）。不同T分期分析结果显示，T1b期患者淋巴结清扫数目1~7枚组患者预后优于≥8枚组（χ 2=4.610， P<0.05）；T2期患者淋巴结清扫数目≥7枚组患者预后优于1~6枚组（χ 2=4.287， P<0.05）；T3+T4期患者淋巴结清扫数目12~15枚组患者预后优于1~11枚组（χ 2=5.007， P<0.05）及≥16枚组（χ 2=10.158， P<0.01）。 结论:淋巴结清扫数目是胆囊癌意向性根治性切除术后患者的独立预后因素。T1b期患者推荐清扫<8枚淋巴结；T2期患者清扫>6枚淋巴结能改善患者预后；T3期以上患者推荐清扫12~15枚淋巴结，清扫≥16枚淋巴结不能改善预后。

abstractsObjective:To examine the role of the number of lymph nodes examined(NLNE) on the prognosis of patients with curatively resected gallbladder carcinoma(GBC).Methods:The clinicopathological data and prognosis of 401 patients with GBC who underwent radical surgery from six institutions of China from January 2013 to December 2017 were analyzed retrospectively. There were 153 males(38.2%) and 248 females(61.8%), with age of (62.0±10.5) years (range: 30-88 years). Fifty-three patients(22.2%) were accompanied by jaundice. All patients underwent radical resection+regional lymphadenectomy.R0 or R1 resection was confirmed by postoperative pathological examination.The different cut-off values of NLNE were determined by the X-tile software, the optimal cut-off values were identified by analyzing the relationship between different cut-off values of NLNE with survival rate. Kaplan-Meier method was used for survival analysis. Univariate and multivariate analysis were implemented respectively using the Log-rank test and Cox proportional hazard model.Results:Among the 401 patients enrolled, 135 cases (33.6%) had lymphatic metastasis, of which 98 cases were in N1 stage(24.4%) and 37 cases were in N2 stage(9.2%).A total of 2 794 NLNE were retrieved, with a median count of 6 (5).The median positive lymph nodes count was 0 (1), and the median positive lymph nodes ratio was 0 (IQR, 0-0.2). Since the 12 and 15 were determined as the cut-off values by X-tile, all patients were divided into three groups of 1-11, 12-15 and ≥16.The 3-year survival rate of the three groups was 45.2%, 74.5%, 12.0% respectively, with statistically significant difference between three groups (χ 2=10.94, P<0.01). The results of multivariate analysis showed that NLNE was an independent prognostic factor for overall survival ( P<0.05). Further analysis was performed specifically for subgroup of T stages. For T1b patients, the prognosis of the NLNE with 1-7 group was significantly better than that of the ≥8 group(χ 2=4.610, P<0.05). For T2 patients, the prognosis of the TLNE ≥7 group was significantly better than that of 1 -6 group (χ 2=4.287, P<0.05). For T3 and T4 patients, the prognosis of the TLNE with 12 - 15 group was significantly better than that of 1 -11 group (χ 2=5.007, P<0.01) and ≥16 group (χ 2=10.158, P<0.01). Conclusions:The NLNE is an independent factor affecting the prognosis of patients with GBC.For patients with stage T1b,8 lymph nodes should be retrieved; for patients with stage T2,extensive dissection of more than 6 lymph nodes can significantly improve the prognosis.For advanced patients (stages T3 and T4), extensive dissection with 12-15 lymph nodes is recommended. However, it fails to get more survival benefits by dissecting more than 16 lymph nodes.